Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/baldaworks/callee/create-agentnpx skills add baldaworks/callee --skill create-agentgit clone --depth 1 https://github.com/baldaworks/calleeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/baldaworks/callee/create-agent)<a href="https://agentmods.dev/skills/baldaworks/callee/create-agent"><img src="https://agentmods.dev/badge/skills/baldaworks/callee/create-agent.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00048 | $0.01844 |
| Opus 5 | $0.00024 | $0.00922 |
| Sonnet 5 | $0.00010 | $0.00369 |
| Haiku 4.5 | $0.00005 | $0.00184 |
Grade A, and why
create-agent scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
This is a copy
97% identical to callee-create-agent — 2 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.
How it starts
The opening of the file, as written. The whole thing — 148 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Create a Callee agent
Use callee when available. Otherwise use the pinned fallback npx --yes @baldaworks/[email protected] for every command in the task.
Choose the kind and ID
Inspect the existing catalog before writing:
callee agent list --json
Choose exactly one supported kind:
Role: one provider-backed leaf agent.Script: one deterministic local validator leaf.Human: one operator-backed interactive leaf.Sequential: ordered child agents that each run once.Loop: ordered child agents repeated until an authorized Role escalates or the iteration limit is exhausted.Router: exactly one named or default child selected by a deterministic route template.
Do not create Parallel. Write below .callee/. Markdown is the default; use a complete .yaml or .yml object only when the user explicitly requests YAML. Directories such as roles/ and workflows/ are optional ID namespaces, not kind selectors. The agent ID is the relative path without its final supported extension.
Choose the authored interaction profile
Choose the profile before authoring any Role or Human. Keep whole-run mode, Role protocol, and ACP permissions distinct:
| Intent | Role fields | Default execution path |
|---|---|---|
| Supervised one-shot (default) | Omit spec.interactive; use permissions.mode: ask |
Interactive whole run for permission prompts; one-shot Role |
| Conversational | Set spec.interactive: true; choose permissions independently |
Interactive whole run and REPL Role |
| Unattended restricted | Omit spec.interactive; use permissions.mode: deny |
Non-interactive when the tree has no Human and all inputs are supplied |
| Unattended approved | Omit spec.interactive; use permissions.mode: allow only on explicit request |
Non-interactive when the tree has no Human and all inputs are supplied |
Preserve the supervised one-shot profile when the user does not specify a
mode. Do not turn ask into allow or deny merely to avoid a controlling
TTY. Use deny for unattended execution without approval to grant
permissions, and use allow only when the user explicitly authorizes automatic
approval. Do not add a Human anywhere in an unattended tree, including beneath
an unselected Router branch.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 148 lines · 48 tokens per session scan A 60d2c693f0d8
create-agent is a skill published in the GitHub repository baldaworks/callee (74 stars, last pushed 25d ago), licensed MIT. It adds 48 tokens to every session and 1,844 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. It is 97% identical to callee-create-agent, differing in 2 lines, and is treated as a copy.
Other skills, from other repositories
agent-communication-protocol
Open protocol for AI agent interoperability enabling standardized communication between agents, applications, and humans across different frameworks.
foblex-flow
Authoritative guide for working with Foblex Flow (@foblex/flow) in the skill-map UI. Use whenever editing Angular code in ui/ that touches graph rendering — templates with f-flow / f-canvas / f-connection / fNode / fConnector / fDraggable / fZoom / fMarker directives; TypeScript importing from @foblex/flow…
researchos-condense
Distill and condense sources into topic knowledge.md along the half-life axis L3→L2→L1→L0. Stable knowledge stays in the KB; fast knowledge prefers live fetch. Core ResearchOS product skill — use after discovery.
researchos-grow
Run one ResearchOS growth cycle: prime from L0/L1 → discover on user-trusted channels → capture → distill/corroborate by half-life → think → emit a lowest-burden user surface (act / hold / flip), audit behind. Use when the user wants to grow/deepen/research a topic.
researchos-search
Source-agnostic evidence gathering for ResearchOS. Use whatever channels the user trusts (APIs, browser, files, briefings, library). Multi-angle coverage matters more than any single transport.
pr-finish-pass
Final-polish pass on the current PR or feature branch before it lands. Rebases on origin/main, force-pushes, drives CI to green, then reviews the diff for correctness/perf/security/reliability bugs, flaky test patterns banned by make lint-test-patterns, sloppy comments a senior engineer would not write, cross-crate…